Купить живой трафик

2 руб / подписчик

Статистика Избранное

Just links

Гео и язык канала: Россия, Английский

Категория: Образование

That's just link aggregator of everything I consider interesting, especially DL and topological condensed matter physics. @EvgeniyZh

Связанные каналы | Похожие каналы

Гео и язык канала

Россия, Английский

Категория

Образование

Статистика

Избранное

Фильтр публикаций

Скрывать удаленные

Скрывать репосты

Just links

6 Oct, 23:27

https://fixupx.com/OREAXEAX/status/2107435564429369632

Or Zamir (@OREAXEAX)

(1/4) Awesome breakthrough from Anthropic+@pinkfloydie+@firebat03! Many other conjectures (e.g., SETH) are very plausibly also false. Nonetheless, here is a short thread about what seems to have made this problem especially "ripe for AI" (in hindsight). https://arxiv.org/abs/2610.06783v1

298 1 1 1 2

Just links

6 Oct, 23:22

https://fixupx.com/percyliang/status/2107502164902031487

Percy Liang (@percyliang)

🚢Marin 535B-A23B training has crossed the halfway mark! Some observations and learnings so far:

316 1 2

Just links

6 Oct, 21:50

Qubit CSS codes with Universal Transversal Gates https://arxiv.org/abs/2610.06730

Qubit CSS codes with Universal Transversal Gates

The Eastin-Knill theorem rules out a universal set of transversal logical gates on any code that detects single-qubit errors. The theorem constrains one partition of the qubits at a time. We show...

407 1 0 1

Just links

5 Oct, 22:13

Planning to Learn https://arxiv.org/abs/2610.03667

Planning to Learn

Policy-gradient methods are central to modern reinforcement learning, including LLM post-training. When they struggle, the usual suspects are exploration, credit assignment and action-sampling...

697 1 21 5

Just links

4 Oct, 07:11

Topological Transverse Transport without a Gap in Critical Topological Flat Bands https://arxiv.org/abs/2609.40065

Topological Transverse Transport without a Gap in Critical...

Quantized Hall transport is traditionally anchored to a bulk spectral gap, which isolates the occupied subspace and exponentially suppresses thermal deviations. Recently discovered critical...

855 1 5 3

Just links

1 Oct, 22:55

Towards verifiable quantum advantage with random circuits: Observables that survive concentration https://arxiv.org/abs/2609.37890

Towards verifiable quantum advantage with random circuits:...

Demonstrating quantum advantage on current quantum hardware is a central goal of quantum computing, and random quantum circuits underpin many leading proposals. Yet sampling-based demonstrations...

1.3k 1 1

Just links

1 Oct, 21:27

Parallel classical simulation of noisy shallow circuits: no quantum advantage in 1D https://arxiv.org/abs/2609.39448

Parallel classical simulation of noisy shallow circuits: no...

We consider quantum circuits consisting of $d$ layers of nearest-neighbor two-qubit gates acting on $n$ qubits arranged on a line, where every qubit is independently depolarized with a constant...

1.2k 1 4 1

Just links

1 Oct, 21:25

https://swesweep.com/

Given a real repository, an agent must discover & repair as many bugs as they can. Agents are not given any hint about the type of bug or its location.

1.1k 1 12 6

Just links

1 Oct, 20:23

Designing Strongly Correlated Quantum Phases of Matter with Foundation Neural-Network Quantum States https://arxiv.org/abs/2609.40209

Designing Strongly Correlated Quantum Phases of Matter with...

Designing a material with a desired property amounts to solving an inverse problem: finding the couplings of a microscopic Hamiltonian whose ground state exhibits that property. For strongly...

1k 1 4 7

Just links

1 Oct, 10:42

Natural Image Pretraining Improves Abstract Reasoning https://link.springer.com/chapter/10.1007/978-3-032-37517-9_6

Natural Image Pretraining Improves Abstract Reasoning

Can visual representations learned from natural images transfer to abstract visual reasoning? We study this question on the Abstraction and Reasoning Corpus (ARC), where each task provides only a few input–output demonstrations and requires inferring a latent...

1.1k 1 11 5

Just links

30 Sep, 20:15

Context Language Models https://arxiv.org/abs/2609.37725

Context Language Models

We introduce Context Language Models (CLMs), language models that natively manage their own context. We implement this by treating the context as a file and allowing the model to make unrestricted...

1.1k 1 29 5

Just links

28 Sep, 22:11

Disentangling the Toric Code https://arxiv.org/abs/2609.30363

Disentangling the Toric Code

Commuting-projector Hamiltonians with integer spectrum such as the 1+1d Ising ferromagnet and the 2+1d toric code may be regarded as the generators of non-on-site $U(1)$ symmetries. In this paper...

1.3k 1 5 1

Just links

28 Sep, 07:11

Spurious probes as a black-box alternative to activation probing https://www.lesswrong.com/posts/gZh6txHhp8sm832sE/spurious-probes-as-a-black-box-alternative-to-activation

Spurious probes as a black-box alternative to activation probing — LessWrong

TL;DR • * We study spurious probes: unrelated questions that reveal internal states of models. Asked "Suggest a type of amphibian." at the end of a…

1.4k 1 8 3

Just links

26 Sep, 14:23

Learning to Discover Interesting Mathematics https://arxiv.org/abs/2609.28603

Learning to Discover Interesting Mathematics

Recently, Large Language Models (LLMs) have been increasingly able to solve advanced mathematical problems, including many that have been open for decades. This opens the door to expansion of...

1.7k 1 25 6 5

Just links

26 Sep, 13:17

https://fixupx.com/syhw/status/2103488919819948513

Видео недоступно для предпросмотра

Смотреть в Telegram

Gabriel Synnaeve (@syhw)

In 2016 I set out to solve Brood War with self-play RL from scratch. Today I believe that this quest has been completed https://www.youtube.com/watch?v=Lfo4C0xzU04 (by a 300M params model).

1.7k 2 10

Just links

26 Sep, 12:19

Self-Play Pretraining with Zero Data https://arxiv.org/abs/2609.30063

Self-Play Pretraining with Zero Data

Advances in language modeling have been driven by scaling pretraining on ever more data. Yet, the training data is still largely curated on the model's behalf. A more general approach to...

10.3k 3 107 10

Just links

24 Sep, 19:02

Cheaters and whistleblowers in the agent swarm https://institute.deepmind.com/essays/cheaters-and-whistleblowers-in-the-agent-swarm

Cheaters and whistleblowers in the agent swarm

In a 100-agent virtual math conference, a cascade of cheating emerged—but then other agents fought back. Ensuring good outcomes is not just a matter of aligning individual models, but designing the right institutions and infrastructure for entire societies of agents.

2.2k 4 29 3

Just links

22 Sep, 02:56

Small-Scale Experiments: Are We There Yet? https://arxiv.org/abs/2608.11859

Small-Scale Experiments: Are We There Yet?

Scaling laws promised cost-effective experiments; six years later, they have yet to fully deliver. Instead, researchers have found them unreliable at small scales (starting at 4M parameters) and...

1.8k 1 27 1 1

Just links

20 Sep, 14:03

Classification of Rational c=1 Vertex Operator Algebras and Vertex Operator Superalgebras https://arxiv.org/abs/2609.00122

Classification of Rational $c=1$ Vertex Operator Algebras and...

For mathematicians: In this first in a series of two papers, we give a mathematically rigorous classification of (sufficiently nice) $c=1$ vertex operator algebras (VOAs) and vertex operator...

1.9k 1 5

Just links

17 Sep, 14:36

Choosing the Trainable Geometry for Agentic RL https://www.trajectory.ai/field-notes/choosing-the-trainable-geometry-for-agentic-rl

Research lab and product company building the platform for continual learning.

2.4k 1 26

Показано 20 последних публикаций.

6 928

подписчиков

Статистика канала

Популярное в канале

Self-Play Pretraining with Zero Data https://arxiv.org/abs/2609.30063

Choosing the Trainable Geometry for Agentic RL https://www.trajectory.ai/field-notes/choosing-the...

https://fixupx.com/qingyu_shi_/status/2100181567666901052

Cheaters and whistleblowers in the agent swarm https://institute.deepmind.com/essays/cheaters-and...

https://fixupx.com/finlay_sanders/status/2099518352536305850